The exponential growth of scientific literature necessitates advanced computational methods to trace conceptual evolution and semantic relationships. This study investigates N-gram structures and concepttransition hierarchies within 96 research articles from the 2025 issues of the journal Polymers. Using NLPbased corpus analysis, we evaluated lexical diversity, phrase density, and collocation strength at the unigram, bigram, and trigram levels. Furthermore, we modelled concept expansion trees, semantic pathway networks, and community structures, validating our findings through rigorous bootstrap and permutation testing. The Results shown a progressive increase in lexical diversity and collocation strength from unigrams to trigrams, confirming a systematic shift toward highly specialised, cohesive technical expressions. Core foundational concepts such as "model" and "thermal" exhibit hierarchical expansion into complex trigrams (e.g., "deep learning model"), highlighting the growing integration of computational intelligence into traditional thermokinetics and materials research. Semantic network analysis reveals a scale free architecture dominated by central hub concepts, while community detection identifies four distinct thematic clusters that facilitate targeted interdisciplinary knowledge transfer. Ultimately, the polymer literature corpus exhibits a mature, cumulative, and internally coherent knowledge system. This integrated methodological approach provides a robust, statistically validated framework for mapping the development of scientific terminology, disciplinary maturity, and interdisciplinary convergence in rapidly evolving research domains, offering valuable insights for future Natural Language Processing.
[1] Huang, Y. C., Lin, H., Hsu, Y. L., et al. (2012). Using n-gram analysis to cluster heartbeat signals. BMC Medical Informatics and Decision Making, 12, Article 64. https://doi.org/10.1186/1472-6947-12-64.
[2] Çiftçi, A., Vural, A., Ural, M. N. (2021). Analysis of some concepts related to the environment and health with the n-gram method. Journal of International Health Sciences and Management, 7(13), 47–54. https://doi.org/ 10.48121/jihsam.796465.
[3] Babur, Ö., Cleophas, L. (2017). Using n-grams for the automated clustering of structural models. In B. Steffen, C. Baier, M. van den Brand, J. Eder, M. Hinchey, T. Margaria (Eds.), SOFSEM 2017: Theory and Practice of Computer Science (Vol. 10139, Lecture Notes in Computer Science). Springer. https://doi.org/10.1007/978-3-319-51963- 0_4.
[4] Sidorov, G. (2019). Syntactic n-grams in computational linguistics. Springer.
[5] Ural, M. N., Tuna, Ö. (n.d.). The analysis of basic concepts related to corporate sustainability by using n-gram analysis technique.
[6] Sabra, S., Sabeeh, V. (2020). A comparative study of n-gram and skip-gram for clinical concepts extraction. In 2020 International Conference on Computational Science and Computational Intelligence (CSCI) (p. 807–812). IEEE. https://doi.org/10.1109/CSCI51800.2020.00151.
[7] Oates, L., Edwards, A., Ersoy, A., van Bueren, E. (2022). A corpus-assisted discourse analysis of sustainability transitions in urban basic infrastructure services. European Journal of Spatial Development, 19(4), 44–71. https:/ /doi.org/10.5281/zenodo.6965763.
[8] Varre, A., Yüce, G., Flammarion, N. (2025). Learning in-context n-grams with transformers: Sub-n-grams are near-stationary points. arXiv. https://arxiv.org/abs/2508.12837.
[9] Svete, A., Cotterell, R. (2024). Transformers can represent n-gram language models. arXiv. https://arxiv.org/ abs/2404.14994.
[10] Rajaraman, N., Bondaschi, M., Ramchandran, K., Gastpar, M., Makkuva, A. V. (2024). Transformers on Markov data: Constant depth suffices. arXiv. https://arxiv.org/abs/2407.17686.
[11] Makkuva, A. V., Bondaschi, M., Girish, A., Nagle, A., Jaggi, M., Kim, H., Gastpar, M. (2024). Attention with Markov: A framework for principled analysis of transformers via Markov chains. arXiv. https://arxiv.org/abs/ 2402.04161.
[12] Bietti, A., Cabannes, V., Bouchacourt, D., Jégou, H., Bottou, L. (2024). Birth of a transformer: A memory viewpoint. Advances in Neural Information Processing Systems, 36.
[13]Nichani, E., Damian, A., Lee, J. D. (2024). How transformers learn causal structure with gradient descent. arXiv. https://arxiv.org/abs/2402.14735.
[14] Edelman, B. L., Edelman, E., Goel, S., Malach, E., Tsilivis, N. (2024). The evolution of statistical induction heads: In-context learning Markov chains. arXiv. https://arxiv.org/abs/2402.11004.
[15] Khaghaninejad, M. S., Eslami, M., Yadollahi, S., Jafari, S. M. (2021). A corpus-based analysis of the application of concluding transition signals in academic texts. Cogent Arts Humanities, 8(1). https://doi.org/10.1080/ 23311983.2020.1868223.
[16] Feola, G., & Jaworska, S. (2019). One transition, many transitions A corpus-based study of societal sustainability transition discourses in four civil society proposals. Sustainability Science, 14, 1643–1656. https://doi.org/ 10.1007/s11625-018-0631-9.
[17] Condamines, A. (2002). Corpus analysis and conceptual relation patterns. Terminology: International Journal of Theoretical and Applied Issues in Specialized Communication, 8(1), 141–162.
[18] Surya, M., Satyawati. (2023). Corpus-based analysis of transition words. CaLLs: Journal of Culture, Arts, Literature, and Linguistics, 189–200. https://e-journals.unmul.ac.id/index.php/CALLS/article/view/13151.
[19] Shibuya, Y., Jensen, K. E. (2015). The functionality of language mining for constructions in texts using n-gram and network analysis. Globe: A Journal of Language, Culture and Communication, 2.